The broad, long-term objective motivating the current proposal is to develop dynamic optical tomography into an economical and accurate clinical tool for real-time functional brain imaging. As an initial demonstration of feasibility, dynamic optical tomography will be used in an attempt to replicate the findings of an earlier study, which employed functional magnetic resonance imaging, of the left posterior regions of the human cerebral cortex. Many earlier functional imaging studies have shown that there are detectable differences between the levels of neuronal activity induced in this region by a purely visual task and by one that also involves reading. It is expected that there will be attendant differences in blood volume and oxygenation between these two tasks. Then comparison of dynamic optical tomographic data and images obtained during the two tasks would a useful test for the feasibility of achieving the long-term objective. The first specific aim of the proposed research is to address the issues of relativity low spatial resolution and accuracy that have hindered the adoption of optical tomography as a clinical imaging modality. A numerical strategy is proposed that is expected to substantially improve spatial resolution and accuracy, in a computationally efficient manner. The second specific aim is to determine whether there are spatially localizable, significant differences between the visual-task and reading-task dynamic optical data taken from participants in a clinical study. It is expected that the mathematical procedure that was optimized pursuant to the previous aim will significantly enhance the ability of dynamic optical tomography to accurately localize regions of increased neuronal activity.